Rounding discrepancy after stock split

I am recording a stock split, 3 for 2. The transfer agent is using 3 decimal places, so 115.983 shares became 173.975 shares. Quicken shows 173.9745 shares. It appears to me that there are 2 possible solutions. 1) Use the adjust share balance feature. 2) Edit stock split, using 173.975 new shares for 115.983 old shares. Any opinions out there about the consequences of either method would be appreciated.

Answers

  • q_lurker
    q_lurker Quicken Windows Subscription SuperUser ✭✭✭✭✭
    I would be inclined to 
    a) ignore it -- let Quicken continue on with the mission 0.0005 shares, or
    b) Enter a Remove Shares followed by and Add Shares bringing the share quantities in line and maintaining the proper cost basis info and acquisition dates.  That could be as Remove 115.983 shares and Add 173.975 shares (no split transaction) or doing it after the split as Remove 173.7545 shares and Add 173.975 shares.

    I would avoid the 173.975 / 115.983 split because I doubt you will end up with the 3-digit precision you are after.  (Try it if you like.)

    I think the Adjust Share Balance is nothing more than a placeholder which I characteristically avoid.

    You may want to watch how this gets handled in the price graph for the Security Details where you can have that shown natural or split-adjusted.  That may bias you to the post-split option b option.
  • D Bunker
    D Bunker Member ✭✭
    Dear q_lurker: As I see it, remove / add shares results in the 'loss' of individual lots, or to put it another way, replacing individual lot cost basis with average cost basis. Furthermore, the holding period appears to be reset to the add date for all lots. Not what I want here.

    The adjust share balance method results in 0.0005 shares with no cost basis, the consequences of which I cannot foresee, but hope are insignificant.

    The 173.975 / 115.983 split worked on a trial 2 lot model. I guess I will back up the real data and attempt this modified split ratio. This is the Lucent arm of the AT&T divestiture with another split (Agere A&B) and 2 mergers to go. 

    Thank goodness for the Stabilization Act of 2008 which requires transfer agents to track cost basis from 2011 forward. Although after a recent return of capital, I would say that compliance is still not 100%. 
  • Tom Young
    Tom Young Quicken Windows Subscription SuperUser ✭✭✭✭✭
    "Although after a recent return of capital, I would say that compliance is still not 100%."

    Completely off-topic but that caught my eye.  Brokers have had a decade to dial in and tweak their cost systems and returns of capital aren't exactly rare, so I'd be surprised that you'd see an error here.  If you don't mind: name of broker and what sort of error you're seeing?

    Since returns of capital aren't completely nailed down until 1099s are issued and then frequently reissued as "corrected" later, maybe it's simply a timing issue?
  • D Bunker
    D Bunker Member ✭✭
    Dear Tom,
    The last time I looked at Windstream Holdings online, I think that the transfer agent had not properly accounted for the reclassification of dividends as a return of capital (or at least not transparently). This occurred many times for this stock, so it was not a timing issue. I was using an Excel spreadsheet (not Quicken) for those calculations, so I will re-run the calculations in Quicken, but not soon as I am slogging my way painfully through AT&T at this time.
  • D Bunker
    D Bunker Member ✭✭
    Dear q_lurker,
    You were right about the failure to get the digit precision I desired! I am 0.000036 shares off with the modified ratio of 1.500004. Is that better than 0.0005? Adjust the share balance for the smaller number? I guess.
    I did not understand your suggestion of looking at "the price graph for the Security Details". Could you explain?
  • q_lurker
    q_lurker Quicken Windows Subscription SuperUser ✭✭✭✭✭
    D Bunker said:
    ... 
    I did not understand your suggestion of looking at "the price graph for the Security Details". Could you explain?
    The security detail view includes a view of the price history timeline (or the market value timeline, but I am interested in the price timeline for this point).



    The raw data should show a drop in the price corresponding to the date of the split (as shown above for a 2:1 split).  Or you can check the Split Adjust box and the 'prior to split' prices will be adjusted accordingly and the graph should take on a continuous more uniform look.  

    If you do not have the stock split recorded in the data as when I suggested a simple Remove Shares / Add Shares approach, you probably won't see the split adjust checkbox make the right adjustment.  Looking at the market value across that time point there is not split adjust option, but you should be ok.
  • q_lurker
    q_lurker Quicken Windows Subscription SuperUser ✭✭✭✭✭
    D Bunker said:
    Dear q_lurker,
    You were right about the failure to get the digit precision I desired! I am 0.000036 shares off with the modified ratio of 1.500004. Is that better than 0.0005? 
    0.000500 vs 0.000036?  I am not sure that is particularly better.  Or is it 14x better?  Your choice.  

    My first and second inclinations remain the same - ignore, or use Remove Shares / Add Shares.  If there are a lot of "Adds" to add, you can do a Shares Transferred action to transfer them from "original account" to the same "original account".  You would then be simply editing the share quantity for each Add to 150% rounded to 3 digits as you saw fit.  The basis and date would already be there. 

    I see nothing really wrong with the 'ignore' route, though at some point down the line you may till need to adjust. for the roundoff. 
  • D Bunker
    D Bunker Member ✭✭
    To quote The Who: "That's my solution. Watch the police and the taxman miss me!"
    I "fixed" the share discrepancy by adding 0.000036 shares to the next reinvested dividend. It makes the math cleaner.
  • q_lurker
    q_lurker Quicken Windows Subscription SuperUser ✭✭✭✭✭
    D Bunker said:
    To quote The Who: "That's my solution. Watch the police and the taxman miss me!"
    I "fixed" the share discrepancy by adding 0.000036 shares to the next reinvested dividend. It makes the math cleaner.
    Seems quite reasonable.